Abstract:
Deterministic models are predominant in atmospheric and climate research
because they represent knowledge about physical processes and because the
parameters have a clear physical interpretation. In this talk I will give
three examples for the role of statistics within the framework of
deterministic models. The first one concerns data assimilation where
observations are used sequentially to correct forecasts. These are then used
as initial conditions for the next forecast. I will discuss and compare two
Monte Carlo methods for this problem, called ensemble and particle filter. The
second example concerns a Bayesian combination of regional climate predictions
from different climate models. There the crucial issue are the assumptions
about the prediction biases. Finally, I will introduce the technique of timevarying parameters to diagnose deficits of deterministic models. This will be
illustrated by a simple hemispherically averaged energy balance model for
global climate.